emotional intelligence

www.elsevier.com/locate/paid

Personality and Individual Differences 42 (2007) 1347–1358

Trait emotional intelligence and decision-related emotions

Nick Sevdalis a,c,*, K.V. Petrides b, Nigel Harvey a

a University College London, United Kingdomb Institute of Education, University of London, United Kingdom

c Clinical Safety Research Unit, Department of Bio-Surgery and Surgical Technology, Imperial College London,

10th floor, QEQM, St. Mary’s Hospital, South Wharf Road, London W2 1NY, United Kingdom

Received 2 May 2006; received in revised form 29 September 2006; accepted 4 October 2006Available online 22 November 2006

Abstract

Although substantial empirical research has emphasized the relevance of emotions in decision-makingprocesses, individual differences in the perception and experience of emotion have been largely overlooked.Here we report research that examines the relationship between trait emotional intelligence (trait EI or trait

emotional self-efficacy) and decision-related affect. In Study 1, we obtained a positive relationship betweentrait EI and the deterioration of mood after the recall of a poor real-life decision. In Study 2, we obtained anegative relationship between trait EI and negative emotions experienced a few days after a failed negoti-ation. In addition, trait EI was positively associated with affective overprediction. The findings are dis-cussed with reference to behavioural decision research and the need to explore the role of individualdifferences in this research.� 2006 Elsevier Ltd. All rights reserved.

Keywords: Regret; Disappointment; Mood; Trait emotional intelligence; Trait emotional self-efficacy; Affectiveself-prediction; TEIQue

0191-8869/$ - see front matter � 2006 Elsevier Ltd. All rights reserved.doi:10.1016/j.paid.2006.10.012

* Corresponding author. Address: Clinical Safety Research Unit, Department of Bio-Surgery and SurgicalTechnology, Imperial College London, 10th floor, QEQM, St. Mary’s Hospital, South Wharf Road, London W21NY, United Kingdom. Tel.: +44 0 207 886 6567; fax: +44 0 207 886 1810.

E-mail address: [email protected] (N. Sevdalis).

mailto:[email protected]

1348 N. Sevdalis et al. / Personality and Individual Differences 42 (2007) 1347–1358

1. Trait emotional intelligence and decision-related emotions

When people make decisions, they often think about the emotions that the outcomes of theirdecisions are likely to trigger. Decision-makers anticipate their emotions before a decision mate-rialises, they experience them when they receive the outcome of their decisions, and they recallthem from memory when they contemplate past decisions (good or poor). Evidence from behav-ioural decision research suggests that the emotions people expect to experience in the future, orthose that they have experienced in the past as a consequence of their decisions, are importantdeterminants of their behaviour (Mellers, Schwartz, & Ritov, 1999).

Research from the individual differences tradition suggests that people differ in systematic waysin how they experience emotions, how able they are to differentiate between them, and how muchemotional information they can process (Winter & Kuiper, 1997). However, individual differencesin the way people manage their decision-emotions have traditionally been neglected by behav-ioural decision researchers.

Assessing individual differences in the management of decision-related emotions requires theutilisation of a construct that operationalizes emotion-related individual differences successfullyand comprehensively. Such an operationalization is provided by the construct of trait emotionalintelligence (trait EI or trait emotional self-efficacy), which comprises a constellation of emotion-related self-perceptions and dispositions located at the lower levels of personality hierarchies (Pet-rides & Furnham, 2001). Unlike other EI approaches, trait EI theory does not view the constructas necessarily adaptive or desirable. For example, Petrides and Furnham (2003) found that hightrait EI participants showed greater mood deterioration than their low trait EI peers followingexposure to distressing stimuli. This is a tenet that makes trait EI theory particularly suitableas an explanatory framework for our purposes.

Other studies that have examined the relationship between EI-related questionnaires and moodinclude Ciarrochi, Chan, and Bajgar (2001) and Schutte, Malouff, Simunek, McKenley, and Hol-lander (2002). These will now be discussed from the perspective of trait EI theory, which may con-tradict the hypotheses and interpretations of the original studies. Interested readers areencouraged to read the original sources. For a discussion of trait EI theory and how it differs fromthe other approaches to emotional intelligence, see Petrides, Furnham, and Mavroveli (in press).

Schutte et al. (2002) measured positive affect (PA) before and after a negative mood inductionthrough Velten (1968) procedure. They found a positive correlation between trait EI and reduc-tion in PA levels, which they interpreted as evidence that ‘‘emotionally intelligent’’ individuals arebetter at regulating their mood. However, two ambiguities render this conclusion premature.First, the appropriate dependent variable in this case was negative, not positive, affect. It is un-clear why, instead of examining the effects of negative mood induction on negative affect (NA)change, the study focused exclusively on the (conceptually and empirically independent in the PA-NAS system) positive component of mood. The second point is that even if a robust effect on NAhad been demonstrated through the Velten procedure, it would lend little credence to the notionthat ‘‘emotional intelligence’’ is a desirable mental ability. Such a conclusion would have requiredus to believe that ‘‘emotionally intelligent’’ participants, in a matter of a few minutes, not onlyread, understood, and genuinely experienced the negative mood states as required by the Veltenmethod, but that they also found the time to regulate them adaptively, a highly unlikelypossibility.

N. Sevdalis et al. / Personality and Individual Differences 42 (2007) 1347–1358 1349

Ciarrochi et al. (2001) study is arguably more illuminating, although somewhat more compli-cated too. They induced negative, neutral, and positive mood through video clips and subse-quently asked participants to look at an ambiguous picture, to generate a short story based onit, and to report how they felt. Their analysis focused on a specific factor of trait EI (viz., emotionregulation), which means the results do not necessarily generalize to the global construct. Amongother things, Ciarrochi et al. found that the group with high emotion regulation scores was morelikely than the group with low scores to generate positive short stories (as rated by independentjudges). However, when the mood ratings of the two groups, which were obtained after the gen-eration of the stories, were compared, they were not significantly different. Moreover, the studyfound that, in the neutral condition, participants with high scores on emotion regulation weremore likely to generate negative stories compared to their peers with low scores.

Our interpretation of existing studies is that they are far indeed from supporting the notion that‘‘emotional intelligence’’ is anything to do with desirable mental abilities, or competencies, orskills. Instead, these studies are generally compatible with the conceptualization of emotionalintelligence as a personality trait, one of whose tenets is that there are situations in which highscores will be associated with maladaptive outcomes (Petrides and Furnham, 2003; Petrides,Furnham, & Mavroveli, in press a). This is a possibility we examine in this paper.

The research that we report here extends experimental investigations of trait EI to the do-main of decision-related emotions. In Study 1, we tested the hypothesis that trait EI is associ-ated with reactivity to decision-related emotional information. In Study 2, we tested thehypothesis that trait EI is associated with more effective management of negative post-decisionalemotions.

2. Study 1

Participants recalled a real-life decision that resulted in a poor outcome. Before and after recall-ing the decision, participants recorded their mood. We expected that mood levels would deterio-rate as a result of recalling the poor decision and, more important, that mood deterioration wouldcorrelate positively with trait EI scores (hypothesis H1). As noted, the hypothesis contradicts apopular view that EI is always adaptive. Trait EI theory explicitly rejects models that view theconstruct as a mental ability or competency or skill and instead reconceptualizes the constructas a constellation of emotion-related self-perceptions (i.e., a personality trait) that cannot be as-sumed to have adaptive value irrespective of circumstance and context (Petrides et al., in press a).

There exist, therefore, instances where higher trait EI is associated with maladaptive outcomes(for example, increased susceptibility to negative mood induction). High trait EI individuals maybe more attuned to affect-laden stimuli than their low trait EI peers and, consequently, more likelyto experience mood deterioration immediately after exposure to distressing stimuli (affect sensitiv-ity hypothesis—H1). While, over time, high trait EI may be conducive to mood regulation, in thepresent experimental design participants were asked to describe their mood immediately after theinduction, which does not allow sufficient time to regulate it.

In addition, we expected that trait EI scores would correlate positively with baseline PA (H2a)and negatively with baseline NA (H2b). Due to the well-being component of the construct, we


would expect high trait EI individuals to experience more positive and less negative affect, onaverage. More generally, in the context of hierarchical structures of affect (Tellegen, Watson, &Clark, 1999) trait EI would be expected to correlate positively with positive affectivity and nega-tively with negative affectivity and to exhibit a strong positive association with the trait pleasant-ness–unpleasantness (or happiness–unhappiness) dimension at the apex of the hierarchy.

3. Method

3.1. Participants

Sixty unpaid students at the University of London volunteered to participate (meanage = 25.24 years, SD = 9.69 years; 43 females, gender data missing for one participant).

3.2. Materials and procedure

Participants were given a short questionnaire containing the experimental instructions andmeasures of affect and trait EI. First, they recorded their current mood using the Positive andNegative Affect Schedule (PANAS; Watson, Clark, & Tellegen, 1988). The PA scale consists ofthe items: attentive, interested, alert, excited, enthusiastic, inspired, proud, determined strong,and active. The NA scale consists of the items: distressed, upset, guilty, scared, hostile, irritable,ashamed, nervous, jittery, and afraid. All items are answered on five-point scales (1 = slightly ornot at all; 5 = very much). The PANAS assesses the activation of positively and negatively valen-ced emotions. Its reliability and construct validity have been demonstrated in a number of studies(Crawford & Henry, 2004; Crocker, 1997; Watson et al., 1988) and the scale has been extensivelyused in research with normal and clinical samples.

Next, participants were instructed to recall, and describe in writing, a decision that led them toexperience intense negative affect. When they were finished, they completed the PA and NA scalesagain.

Finally, participants completed the short form of the Trait Emotional Intelligence Question-naire (TEIQue-SF; Petrides & Furnham, 2006). The TEIQue-SF was designed as a measure ofglobal trait EI, although it is also possible to derive from it scores on the four trait EI factors(‘emotionality,’ ‘sociability,’ ‘self-control,’ and ‘well-being’). Its internal consistency is invariablyhigh (>.80) and validity evidence can be found in Petrides and Furnham (2006). As regards itsoverlap with the higher-order personality dimensions, although we are not aware of any relevantanalyses, we would expect it to be similar to that of the long form of the inventory, viz., about70% (Petrides et al., in press a).

In relation to incremental validity, trait EI can account for variance over and above the basicpersonality factors (for a detailed discussion and empirical demonstration, see Petrides, Perez-Gonzalez, & Furnham, in press). Specifically with respect to the incremental validity of the TEI-Que-SF, Furnham and Petrides (2003) found that, while it could account for happiness varianceover and above the NEO-FFI, the latter (as a block of variables) did not have a statistically sig-nificant effect in the presence of the former.


4. Results and discussion

The mean trait EI score was 146.97 (SD = 18.94; median = 146) and the a was .83. The PANASscales also exhibited satisfactory internal consistency (average a of the two administrations:PA = .90 and NA = .87).

In order to investigate the relationship between trait EI and emotional reactivity to the decisionrecall task, we subtracted the PA (NA) scores before the recall task from PA (NA) scores after thetask. This algebraic manipulation resulted in a reactivity index for PA (NA), on which positivescores indicate increases in positive (negative) affect.

The first two grand columns of Table 1 summarise PA and NA scores across the two moodassessments. Participants’ mood deteriorated after they recalled their poor decision, leading tolower PA (t(59) = 3.62, p < .01) and higher NA (t(59) = 3.58, p < .01) after the recall. The thirdgrand column of Table 1 summarises the two reactivity indices. The index is negative for PA, indi-cating a decrease, and positive for NA, indicating an increase.

Table 2 summarises the intercorrelations of the variables in the study. Of particular interesthere is the first column, which shows the correlations between trait EI, PA and NA (beforeand after the decision recall) and the two reactivity indices. As hypothesized, trait EI correlatednegatively with the reactivity index for PA and positively with the reactivity index for NA. In hightrait EI scorers, PA decreased more and NA increased more after recalling a past decision that ledto intense negative affect, which supports H1. Table 2 also shows that trait EI correlated positivelywith baseline PA and negatively with baseline NA, which supports H2a and H2b.

The relationship between trait EI and the two reactivity indices may be affected by a statisticalartefact. Specifically, it could be argued that the resulting correlations are partly due to the higherbaseline positive (and lower baseline negative) affect of high trait EI participants compared to their

Table 1Means and SDs for key variables in Study 1

Before decision recall After decision recall Reactivity index

M SD M SD M SD

Positive affect 27.06 8.28 24.05 8.18 �3.01 6.45Negative affect 15.28 6.01 17.83 6.41 2.55 5.52

Table 2Pearson correlations for key variables in Study 1

Trait EI PA before PA after NA before NA after Reactivity PA Reactivity NA

Trait EI –PA before .29a

PA after .05 .69b –NA before �.41b .15 .27a –NA after �.12 .18 .10 .61b –Reactivity PA �.30a �.41b .38b .14 �.10 –Reactivity NA .32a .04 �.18 �.39b .50b �.28a –

a p < .05.b p < .01; N = 60.


low trait EI peers, since, in the former, PA drops from higher and NA rises from lower than in thelatter. However, this argument is conceptually problematic because high trait EI is associated withbetter baseline mood prior to any experimental manipulation (see H2; Petrides and Furnham, 2003;Petrides et al., in press a). We tested the hypothesis that high trait EI participants will exhibit great-er mood deterioration than their low trait EI peers, irrespective of the underlying mechanism, whichmay well involve them coming down from a higher mood level. Furthermore and relatedly, we didnot seek to demonstrate that the relationship between trait EI and mood change is incremental overprior mood. What our study shows is that trait EI is a positive predictor of mood deterioration,following exposure to negative mood induction. Whether it remains so after higher-order person-ality traits or prior mood are partialled out is an altogether different question and one of little the-oretical importance in trait EI theory (for a discussion of this issue and relevant empiricaldemonstrations, see Petrides & Furnham, 2003; Petrides, Perez-Gonzalez, & Furnham, in press b).

Although in Study 1 we used a real-life decision in the experimental design, we did not examineparticipants’ decisions as they unfolded. Consequently, we were only able to observe the impact ofa past poor decision on current mood. In Study 2, we set out to investigate the relationship betweentrait EI and decision-related affect in a decision that unfolded during the course of the experiment.

5. Study 2

Participants negotiated for money and were asked to predict how they would feel immediatelyafter the negotiation as well as five days later (both in case they succeeded and in case they failed).As part of the experimental procedure, all negotiations failed. Participants recorded their regretand disappointment immediately after the negotiation and also five days later. (The study waspart of a larger experimental session, the details of which we report in Sevdalis & Harvey, submit-ted for publication.)

Based on existing evidence that people typically overstate their affective self-predictions (Loewen-stein & Schkade, 1999), we expected that, overall, participants would give exaggerated predictions oftheir negative emotions immediately after the negotiation as well as five days later (H1). However, wewere also interested to explore whether the extent of overprediction varies as a function of trait EI.

In addition, we expected that trait EI would correlate negatively with regret and disappoint-ment measured both immediately after the failure of the negotiation (H2a) as well as five days later(H2b). These hypotheses stem from the assumption that, due to their positive self-perceptions,high trait EI individuals should be more resilient to stress and failure. Note that this hypothesisis unrelated to the affect sensitivity hypothesis we tested in Study 1, since it concerns resilience toan experienced failure, rather than sensitivity to emotional information.

6. Method

6.1. Participants

Twenty-four individuals volunteered for the study (mean age = 22.21 years; SD = 2.81 years;14 females) and were compensated for their time.


6.2. Materials and procedure

Participants were given a short booklet containing the instructions for the negotiation task andthe measures of affect and trait EI. The study ran in five stages.

In the first stage, participants received the instructions. We used the ultimatum gameas the nego-tiation paradigm. In the simplest form of an ultimatum game (Guth, Schmittberger, & Schwarz,1982), a proposer receives money from the experimenter, which she then has to share with a re-sponder. The only exchange between the negotiators involves the offer that the proposer makesto the responder. If the offer is accepted, the negotiators share the money according to the offer.If it is rejected, neither receives any amount. In this study, all participants were assigned the roleof the proposer and were told they would negotiate with a responder, who was in a different room.

In the second stage, participants predicted their post-negotiation regret and disappointment forimmediately after the task and for a week later. These two emotions are commonly used to assessdecision-related affect. Participants recorded their responses on five-point scales (1 = slightly ornot at all; 5 = very much). Since participants did not know at this stage the outcome of their of-fers, they predicted their emotions under scenarios of both success and failure. Therefore, we ob-tained affective predictions for two outcome contingencies (success versus failure) and two timeperspectives (immediately versus a week later).

In the third stage, participants completed the TEIQue-SF.In the fourth stage, participants received the outcomes of the negotiation. All offers were re-

jected. After receiving this information, participants recorded their actual regret anddisappointment.

In the final stage of the study, participants returned to the laboratory five days later and re-ported their regret and disappointment again.

7. Results and discussion

The mean trait EI score was 149.85 (SD = 19.13; median = 156.50) and the a was .86. Consis-tent with existing research on ultimatum negotiation, participants offered an average of £4.67(SD = £0.92; median = £5) out of the £10 allocated to them by the experimenter. Eighteen outof the 24 participants made offers of £5 or more.

Table 3 summarises the average regret and disappointment predicted and experienced immedi-ately after the negotiation as well as five days later. Table 4 shows the correlations between traitEI and affective predictions and experiences. As can be seen in that table, the correlations between

Table 3Means and SDs for the affective predictions and experiences in Study 2

Predictions Experiences

Regret Disappointment Regret Disappointment

M SD M SD M SD M SD

Immediately 2.25 1.26 3.79 0.78 1.54 0.93 2.88 1.03Five days later 1.87 0.97 2.39 0.89 1.61 0.66 1.78 0.85

Table 4Pearson correlations for key variables in Study 2

Time Trait EI Predictions Experiences

Immediate 5 days later Immediate 5 days later

Regret Disappoint-ment




Predictions Immediate Regret �.13 –Disappointment .00 .23 –

5 days later Regret .03 .83b .08 –Disappointment �.44a �.29 .44a �.15 –

Experiences Immediate Regret �.22 .32 .16 .08 �.42a –Disappointment .03 .09 .45a �.02 .25 .07 –

5 days later Regret �.62b .16 .35 �.08 �.27 .56b .03 –Disappointment �.49a �.04 �.14 �.20 �.18 .26 .32 .49a –

a p < .05.b p < .01; N = 23–24.

1354N

.S

evda

liset

al.

/P

erson

ality

an

dIn

divid

ua

lD

ifferen

ces4

2(

20

07

)1

34

7–

13

58


trait EI and immediate affect did not reach significance, thus failing to support H2a. In contrast,the correlations with regret and disappointment experienced five days after the negotiation wereboth significantly negative, thus supporting H2b.

Next, we calculated prediction error indices for regret and disappointment by subtracting affec-tive predictions from experiences (both for immediate judgments and for judgments made fivedays later). Negative signs on these indices indicate overprediction (prediction > experience),whereas positive signs indicate underprediction (prediction < experience). We submitted the errorindices to one-sample t-tests against a test value of zero, which indicates perfectly accurate predic-tions (prediction = experience). As hypothesised, we obtained significant overpredictions for re-gret (M = �0.71, SD = 1.30; t(23) = 2.67, p < .05) and disappointment (M = �0.92, SD = 0.97;t(23) = 4.61, p < .001) experienced immediately after the negotiation. As regards judgments madefive days after the negotiation, the results were consistent for disappointment (M = �0.61,SD = 1.34; t(22) = 2.18, p < .05), but not for regret, which did not reach significance(M = �0.26, SD = 1.21; t(22) = 1.03, p = .31). These findings provide support for hypothesis H1.

Trait EI was negatively correlated with the prediction error indices for disappointment(r = �.61, p < .01) and regret (r = �.36, p < .05, one-tailed) five days after the failed negotiation.That is to say, high trait EI participants showed greater discrepancies between predicted and expe-rienced affect than their low trait EI peers a week after the failed negotiation. In contrast, therewas no association between trait EI and affective prediction errors immediately after thenegotiation.

To summarize, Study 2 provided empirical support for two out of the three hypotheses. Partic-ipants overpredicted their regret and disappointment immediately after the unsuccessful negotia-tion and their disappointment five days later (H1). Trait emotional self-efficacy qualified thisfinding, however, with high trait EI participants, more so than their low trait EI counterparts,overpredicting the regret and disappointment that they experienced five days after the failed nego-tiation. High trait EI participants also felt better overall than their low trait EI peers five daysafter the failed negotiation (H2b). An important advantage of the study is that the results wereobtained through a realistic two-person negotiation in which participants were financially moti-vated to perform well as in real-life decision-making.

Study 2 did not find a relationship between trait EI and participants’ affect immediately afterthe negotiation, thus failing to support H2a. As further discussed below, the experimental litera-ture on trait EI is still at an early stage and it is not yet clear which effects are robust nor do weknow much about the processes underlying the various significant effects. In general, we wouldexpect that high trait EI individuals will show greater resilience to stress and failure, an expecta-tion that has been confirmed in several studies (see Mikolajczak, Petrides, Luminet, & Coumans,submitted for publication). It should be reiterated here that the resilience hypothesis is differentfrom the affect sensitivity hypothesis we tested in Study 1.

There is a pressing need for more experimental studies to elucidate the nature and extent of traitEI effects as well as the circumstances under which they become salient. At the same time, it will beimportant to vary systematically the research designs and criteria in order to accumulate evidencethat can inform theory. At present, there are far too many different designs and dependent vari-ables to be able to draw firm conclusions. In the case at hand, the lack of support for H2a could bea function of using highly specific criteria (regret and disappointment) or of measuring themthrough a single item, or of a weak experimental stimulus. In relation to specificity, whereas in


Study 1 we focused on general affective states (i.e., general mood), in Study 2 we assessed decision-specific reactions (i.e., regret and disappointment), which is another reason why the two studiesmay not be directly comparable (see also Connolly, Ordonez, & Coughlan, 1997).

8. General discussion

This paper documents influences of trait EI on decision-related emotions. In Study 1, trait EIwas associated with mood deterioration following the recall of a poor prior decision. In Study 2, itwas associated with larger overpredictions of post-decisional emotions and with a more positiveemotional state five days after receiving a poor decision outcome.

Although these findings must be replicated in larger samples, they suggest that individual dif-ferences in trait emotional self-efficacy influence the anticipation and experience of post-decisionalaffect. Behavioural decision research has traditionally endeavoured to map a variety of environ-mental (i.e., external to the decision-maker) influences on decisions. For instance, it has beenshown that anticipated regret is exacerbated when people expect to receive feedback on the out-come of alternatives they did not choose (Zeelenberg, 1999) and that experienced regret is moreintense for decision-makers who are personally responsible for their decisions than for those whoare not (Zeelenberg, van der Pligt, & de Vries, 2000).

Our research suggests that the above findings may require qualifications. Personal (i.e., internalto the decision-maker) factors might be at work when people contemplate possible routes of ac-tion. For instance, consistent with the existing literature, in Study 2 we found that post-decisionalnegative emotions were overpredicted. However, high trait EI participants overpredicted theseemotions more than their low trait EI peers. Thus, the general pattern of affective overpredictionof decision-related affect was qualified by trait EI.

Which individual differences variables should be targeted by behavioural decision researchers?Trait EI is an obvious candidate, since it provides comprehensive coverage of emotion-relatedself-perceptions that are directly relevant to the study of affective decision-making. Otherresearchers have found that individual differences in the way people weigh the future conse-quences of their decisions are relevant to the experience of post-decisional emotions (Strathman,Gleicher, Boninger, & Edwards, 1994). Such findings illustrate the necessity of exploring ways tointegrate experimental and individual differences approaches to the study of the interrelationshipsbetween decision-making, emotions, and personality.

Our findings are also relevant to trait EI theory (Petrides et al., in press a) and support the viewthat high trait EI is not always adaptive; rather, its adaptive value will vary depending on the con-text. This is a basic tenet of trait EI theory, which differentiates it from approaches that view theconstruct as a cognitive ability, competency, or skill. While a lot of work remains to be done inorder to elucidate the conditions and contexts under which high scores on EI questionnaires havemaladaptive implications, we believe there is sufficient evidence to accept that such conditions andcontexts exist, a fact that only trait EI theory can successfully accommodate.

There are now several studies suggesting that individual differences in trait EI influence howpeople respond to affect-laden emotional stimuli. However, the variability in experimental para-digms prevents us from drawing firm conclusions regarding underlying processes. This is an areawhere further research, with a specific emphasis on careful replication, may be fruitfully pursued.


Acknowledgement

This research was funded in part by the ESRC Centre for Economic Learning and Social Evo-lution and in part by the Greek State Scholarship Foundation (IKY).

References

Ciarrochi, J. V., Chan, A. Y. C., & Bajgar, J. (2001). Measuring emotional intelligence in adolescents. Personality and

Individual Differences, 31, 1105–1119.Connolly, T., Ordonez, L. D., & Coughlan, R. (1997). Regret and responsibility in the evaluation of decision outcomes.

Organizational Behavior and Human Decision Processes, 70, 73–85.Crawford, J. R., & Henry, J. D. (2004). The positive and negative affect schedule (PANAS): Construct validity,

measurement properties and normative data in a large non-clinical sample. British Journal of Clinical Psychology, 43,245–265.

Crocker, P. R. E. (1997). A confirmatory factor analysis of the positive and negative affect schedule (PANAS) with ayoung sport sample. Journal of Sport and Exercise Psychology, 19, 91–97.

Furnham, A., & Petrides, K. V. (2003). Trait emotional intelligence and happiness. Social Behavior and Personality, 31,815–824.

Guth, W., Schmittberger, R., & Schwarz, B. (1982). An experimental analysis of ultimatum bargaining. Journal of

Economic Behavior and Organization, 3, 367–388.Loewenstein, G., & Schkade, D. A. (1999). Wouldn’t it be nice? Predicting future feelings. In D. Kahneman, E. Diener,

& N. Schwarz (Eds.), Well-Being: The Foundations of Hedonic Psychology (pp. 85–105). New York: Russell SageFoundation.

Mellers, B. A., Schwartz, A., & Ritov, I. (1999). Emotion-based choice. Journal of Experimental Psychology: General,

128, 1–14.Mikolajczak, M., Petrides, K. V., Luminet, O., & Coumans, N. (submitted for publication). An experimental

investigation of the moderating effects of trait emotional intelligence on laboratory-induced stress.Petrides, K. V., & Furnham, A. (2001). Trait emotional intelligence: Psychometric investigation with reference to

established trait taxonomies. European Journal of Personality, 15, 425–448.Petrides, K. V., & Furnham, A. (2003). Trait emotional intelligence: Behavioural validation in two studies of emotion

recognition and reactivity to mood induction. European Journal of Personality, 17, 39–57.Petrides, K. V., & Furnham, A. (2006). The role of trait emotional intelligence in a gender-specific model of

organizational variables. Journal of Applied Social Psychology, 36, 552–569.Petrides, K. V., Furnham, A., & Mavroveli, S. (in press a). Trait emotional intelligence: Moving forward in the field of

EI. In G. Matthews, M. Zeidner, & R. Roberts (Eds.), Emotional intelligence: Knowns and unknowns (Series inAffective Science). Oxford: Oxford University Press.

Petrides, K. V., Perez-Gonzalez, J. C., & Furnham, A., (in press b). On the criterion and incremental validity of traitemotional intelligence. Cognition and Emotion.

Schutte, N. S., Malouff, J. M., Simunek, M., McKenley, J., & Hollander, S. (2002). Characteristic emotionalintelligence and emotional well-being. Cognition and Emotion, 16, 769–785.

Sevdalis, N., & Harvey, N., (submitted for publication). Biased forecasting of post-decisional affect.Strathman, A., Gleicher, F., Boninger, D. S., & Edwards, C. S. (1994). The consideration of future consequences:

Weighing immediate and distant outcomes of behaviour. Journal of Personality and Social Psychology, 66, 742–752.Tellegen, A., Watson, D., & Clark, L. A. (1999). On the dimensional and hierarchical structure of affect. Psychological

Science, 10, 297–303.Velten, E. (1968). A laboratory task for induction of mood states. Behavior Research and Therapy, 6, 473–482.Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures of positive and negative

affect: The PANAS scales. Journal of Personality and Social Psychology, 54, 1063–1070.


Winter, K. A., & Kuiper, N. A. (1997). Individual differences in the experience of emotions. Clinical Psychology Review,

17, 791–821.Zeelenberg, M. (1999). Anticipated regret, expected feedback and behavioural decision-making. Journal of Behavioral

Decision Making, 12, 93–106.Zeelenberg, M., van der Pligt, J., & de Vries, N. K. (2000). Attributions of responsibility and affective reactions to

decision outcomes. Acta Psychologica, 104, 303–315.

emotional intelligence

Documents